Hello Windows Insiders, today we’re releasing Windows 11 Build 22631.6269 (KB 5070312) to Insiders in the Release Preview Channel on Windows 11, version 23H2 (Build 22631).   This update includes the following features and improvements that are rolling out as part of this update. Text bolded in brackets indicates the area of the change being documented. This non-security update includes quality improvements. The following summary outlines key issues addressed by the KB after you install it. The bold text within the brackets indicates the item or area of the change.

• [Country and Operator Settings Asset (COSA)]
  • Fixed: This update brings profiles up to date for certain mobile operators.
• [File Explorer]
  • Fixed: This update addresses an issue where File Explorer sometimes didn’t respond to mouse clicks until you closed and reopened it.
  • Fixed: This update addresses an issue with extracting .tar files when file or folder names contain more than 34 commonly used Chinese characters.
• [Group Policy and Configuration]
  • Fixed: This update addresses an issue where the HideRecommendedSection policy didn’t work in Windows 11 Enterprise multi-session environments, such as Azure Virtual Desktop (AVD). Even when configured using Group Policy or Configuration Service Provider (CSP), recommendations still appeared in AVD sessions.

Thanks, Windows Insider Program Team

The pace of AI innovation is accelerating, and developers—across startups and global enterprises—are at the heart of this transformation. Today marks a significant moment for enterprise AI innovation: Azure AI Foundry is unveiling OpenAI’s GPT-5.1 series, the next generation of reasoning, analytics, and conversational intelligence.

The following models will be rolling out in Foundry today:

• GPT-5.1: adaptive, more efficient reasoning
• GPT-5.1-chat: chat with new chain-of-thought for end-users
• GPT-5.1-codex: optimized for long-running conversations with enhanced tools and agentic workflows
• GPT-5.1-codex-mini: a compact variant for resource-constrained environments

What’s new with GPT-5.1 series

The GPT-5.1 series is built to respond faster to users in a variety of situations with adaptive reasoning, improving latency and cost efficiency across the series by varying thinking time more significantly. This, combined with other tooling improvements, enhanced stepwise reasoning visibility, multimodal intelligence, and enterprise-grade compliance.  

GPT-5.1: Adaptive and Efficient Reasoning

GPT-5.1 is the mainline model engineered to deliver adaptive, stepwise reasoning that adjusts its approach based on the complexity of each task. Core capabilities included:

• Adaptive reasoning for nuanced, context-aware thinking time
• Multimodal intelligence: supporting text, image, and audio inputs/outputs
• Enterprise-grade performance, security, and compliance

This model’s flexibility empowers developers to tackle a wide spectrum of tasks—from simple queries to deep, multi-step workflows for enterprise-grade solutions. With its ability to intelligently balance speed, cost, and intelligence, GPT-5.1 sets a new standard for both performance and efficiency in AI-powered development.

GPT-5.1-chat: Elevating Interactive Experiences with Smart, Safe Conversations

GPT-5.1-chat powers fast, context-aware chat experiences with adaptive reasoning and robust safety guardrails. With chain-of-thought added in the chat for the first time, it brings an interactive experience to the next level. It’s tuned for safety and instruction-following, making it ideal for customer support, IT helpdesk, HR, and sales enablement. Multimodal chat (text, image, and audio) improves long-turn consistency for real problem solving, delivering brand-aligned, safe conversations, and supporting next-best-action recommendations.

GPT-5.1-codex and GPT-5.1-codex-mini: Frontier Models for Agentic Coding

GPT-5.1-codex builds on the foundation set by GPT-5-codex, advancing developer tooling with:

• Enhanced reasoning frameworks for stepwise, context-aware code analysis and generation; plus
• Enhanced tool handling for certain development scenario's
• Multimodal intelligence for richer developer experiences when coding

With Foundry’s enterprise-grade security and governance, GPT-5.1-codex is ideal for automated code generation and review, accelerating development cycles with intelligent code suggestions, refactoring, and bug detection.

GPT-5.1-codex-mini is a compact, efficient variant optimized for resource-constrained environments. It maintains near state-of-the-art performance, multimodal intelligence, and the same safety stack and tool access as GPT-5.1-codex, making it best for cost-effective, scalable solutions in education, startups, and cost-conscience settings.

Together, these Codex models empower teams to innovate faster and with greater confidence.

Selecting Your AI Engine: Match Model Strengths to Your Business Goals

One of the advantages of the GPT-5.1 series is unified access to deep reasoning, adaptive chat, and advanced coding—all in one place. Here’s how to match model strengths to your needs:

• Opt for GPT-5.1 for general ai application use—tasks like analytics, research, legal/financial review, or consolidating large documents and codebases. It’s the model of choice for reliability and high-impact outputs.
• Go with GPT-5.1-chat for interactive assistants and product UX, especially when adaptive reasoning is required for complex cases.  Reasoning hints and adaptive reasoning help with customer latency perception.
• Leverage GPT-5.1-codex for deep, stepwise reasoning in complex code generation, refactoring, or multi-step analysis—ideal for demanding agentic workflows and enterprise automation.
• Utilize GPT-5.1-codex-mini for efficient, cost-effective coding intelligence in broad-scale deployment, education, or resource-constrained environments—delivering near-mainline performance in a compact model.

Deployment and Pricing 

Start Building Today

The GPT-5.1 series is now available in Foundry Models. Whether you’re building for enterprise, small and medium-sized business, or launching the next digital-native app, these models and the Foundry platform are designed to help you innovate faster, safer, and at scale.

 

 

---

In this episode of The CSS Podcast, Una and Bramus cover building customizable select menus. Have you ever had to build a dropdown menu where you want to do something as simple as change the color, or add little flag icons? You know how hard it can be! Discover how the web platform is solving this once and for all with the new customizable select API.

Resources:
Customizable select demos → https://goo.gle/43G5ruv 
 
Una Kravets (co-host)
Bluesky | Twitter | YouTube | Website
Making the web more colorful @googlechrome 
Bramus Van Damme (co-host)
Bluesky | Mastodon | YouTube | Website
@GoogleChrome CSS DevRel; @CSSWG; Scuba Diver 🤿

Download audio: https://traffic.libsyn.com/secure/thecsspodcast/TCP098_v1.mp3?dest-id=1891556

NotebookLM adds "Deep Research" for discovering sources, support for Word and PDFs, and a preview of Sheets.